No abstract

Soil moisture plays a key role in the Earth’s water and carbon cycles, but acquisition of continuous (i.e., gap-free) soil moisture measurements across large regions is a challenging task due to limitations of currently available point measurements. Satellites offer critical information for soil moisture over large areas on a regular basis (e.g., European Space Agency Climate Change Initiative (ESA CCI), National Aeronautics and Space Administration Soil Moisture Active Passive (NASA SMAP)); however, there are regions where satellite-derived soil moisture cannot be estimated because of certain conditions such as high canopy density, frozen soil, or extremely dry soil. We compared and tested three approaches, ordinary kriging (OK), regression kriging (RK), and generalized linear models (GLMs), to model soil moisture and fill spatial data gaps from the ESA CCI product version 4.5 from January 2000 to September 2012, over a region of 465,777 km2 across the Midwest of the USA. We tested our proposed methods to fill gaps in the original ESA CCI product and two data subsets, removing 25% and 50% of the initially available valid pixels. We found a significant correlation (r = 0.558, RMSE = 0.069 m3m−3) between the original satellite-derived soil moisture product with ground-truth data from the North American Soil Moisture Database (NASMD). Predicted soil moisture using OK also had significant correlation with NASMD data when using 100% (r = 0.579, RMSE = 0.067 m3m−3), 75% (r = 0.575, RMSE = 0.067 m3m−3), and 50% (r = 0.569, RMSE = 0.067 m3m−3) of available valid pixels for each month of the study period. RK showed comparable values to OK when using different percentages of available valid pixels, 100% (r = 0.582, RMSE = 0.067 m3m−3), 75% (r = 0.582, RMSE = 0.067 m3m−3), and 50% (r = 0.571, RMSE = 0.067 m3m−3). GLM had slightly lower correlation with NASMD data (average r = 0.475, RMSE = 0.070 m3m−3) when using the same subsets of available data (i.e., 100%, 75%, 50%). Our results provide support for using geostatistical approaches (OK and RK) as alternative techniques to gap-fill missing spatial values of satellite-derived soil moisture.

Soil organic carbon (SOC) information is fundamental for improving global carbon cycle modeling efforts, but discrepancies exist from country‐to‐global scales. We predicted the spatial distribution of SOC stocks (topsoil; 0–30 cm) and quantified modeling uncertainty across Mexico and the conterminous United States (CONUS). We used a multisource SOC dataset (>10 000 pedons, between 1991 and 2010) coupled with a simulated annealing regression framework that accounts for variable selection. Our model explained ~50% of SOC spatial variability (across 250‐m grids). We analyzed model variance, and the residual variance of six conventional pedotransfer functions for estimating bulk density to calculate SOC stocks. Two independent datasets confirmed that the SOC stock for both countries represents between 46 and 47 Pg with a total modeling variance of ±12 Pg. We report a residual variance of 10.4 ±5.1 Pg of SOC stocks calculated from six pedotransfer functions for soil bulk density. When reducing training data to define decades with relatively higher density of observations (1991–2000 and 2001–2010, respectively), model variance for predicted SOC stocks ranged between 41 and 55 Pg. We found nearly 42% of SOC across Mexico in forests and 24% in croplands, whereas 31% was found in forests and 28% in croplands across CONUS. Grasslands and shrublands stored 29 and 35% of SOC across Mexico and CONUS, respectively. We predicted SOC stocks >30% below recent global estimates that do not account for uncertainty and are based on legacy data. Our results provide insights for interpretation of estimates based on SOC legacy data and benchmarks for improving regional‐to‐global monitoring efforts.

The current availability of soil moisture data over large areas comes from satellite remote sensing technologies (i.e., radar-based systems), but these data have coarse resolution and often exhibit large spatial information gaps. Where data are too coarse or sparse for a given need (e.g., precision agriculture), one can leverage machine-learning techniques coupled with other sources of environmental information (e.g., topography) to generate gap-free information and at a finer spatial resolution (i.e., increased granularity). To this end, we develop a spatial inference engine consisting of modular stages for processing spatial environmental data, generating predictions with machine-learning techniques, and analyzing these predictions. We demonstrate the functionality of this approach and the effects of data processing choices via multiple prediction maps over a United States ecological region with a highly diverse soil moisture profile (i.e., the Middle Atlantic Coastal Plains). The relevance of our work derives from a pressing need to improve the spatial representation of soil moisture for applications in environmental sciences (e.g., ecological niche modeling, carbon monitoring systems, and other Earth system models) and precision agriculture (e.g., optimizing irrigation practices and other land management decisions).

No abstract

To trust findings in computational science, scientists need workflows that trace the data provenance and support results explainability. As workflows become more complex, tracing data provenance and explaining results become harder to achieve. In this paper, we propose a computational environment that automatically creates a workflow execution's record trail and invisibly attaches it to the workflow's output, enabling data traceability and results explainability. Our solution transforms existing container technology, includes tools for automatically annotating provenance metadata, and allows effective movement of data and metadata across the workflow execution. We demonstrate the capabilities of our environment with the study of SOMOSPIE, an earth science workflow. Through a suite of machine learning modeling techniques, this workflow predicts soil moisture values from the 27 km resolution satellite data down to higher resolutions necessary for policy making and precision agriculture. By running the workflow in our environment, we can identify the causes of different accuracy measurements for predicted soil moisture values in different resolutions of the input data and link different results to different machine learning methods used during the soil moisture downscaling, all without requiring scientists to know aspects of workflow design and implementation.

No abstract

ABSTRACT:Change detection is one of the most important and widely requested applications of terrestrial remote sensing. Despite a wealth of techniques and successful studies, there is still a need for research in remote sensing science. This paper addresses two important issues: the temporal and spatial scales of change maps. Temporal scales relate to the time interval between observations for successful change detection. We compare annual change detection maps accumulated over five years against direct change detection over that period. Spatial scales relate to the spatial resolution of remote sensing products. We compare fractions from 30m Landsat change maps to 250m grid cells that match MODIS change products. Results suggest that change detection at annual scales better detect abrupt changes, in particular those that do not persist over a longer period. The analysis across spatial scales strongly recommends the use of an appropriate analysis technique, such as change fractions from fine spatial resolution data for comparison with coarse spatial resolution maps. Plotting those results in bi-dimensional error space and analyzing various criteria, the "lowest cost", according to a user defined (here hyperbolic) cost function, was found most useful. In general, we found a poor match between Landsat and MODIS-based change maps which, besides obvious differences in the capabilities to detect change, is likely related to change detection errors in both data sets.